WTH, is it normal that the sensitivity of the AF points depends not only on the maximal aperture, but also on the lens type?
Scroll to "AF tracking and ranging patterns can be used for the lens used" on the following Google-translated page to see this...
http://translate.google.com/translate?sl=auto&tl=en&js=n&prev=_t&a mp;ie=UTF-8&layout=2&eotf=1&u=http%3A%2F%2Fcweb. canon.jp%2Fcamera%2Feosd%2F1dx%2Ffeature-afae.html&act=ur
Originally Posted by Colin500
Normal across the board, no. 'Normal' (or at least, precedented) for the 1-series line, yes. From Bryan's review of the 1D IV:
<p style="padding-left: 30px;"]<span>The ... Canon EOS 1D Mark IV features 39 cross-type AF sensors that are functional during manual AF point selection (19 are cross-type functional with automatic AF point selection) with all f/2.8 and faster EF lenses as well as the following f/4 EF lenses:
17-40mm f/4 L USM
24-105mm f/4 L IS USM
70-200mm f/2.8 L IS USM + Extender EF 1.4x II
200mm f/2 L IS USM + Extender EF 2x II
300mm f/2.8 L IS USM + Extender EF 1.4x II
400mm f/2.8 L IS USM + Extender EF 1.4x II
<span>So, the f/2.8 sensor works with some f/4 lenses and TC combos, but not others.
<span>It is interesting (and unusual) from the link you provided for the 1D X is that despite being an f/2.8 lens, the 100mm f/2.8 Macro lenses (both L and non-L) do not activate any of the f/2.8 diagonal cross-type sensors.
BTW, thanks for the link - it provides some details about the AF system that I was looking for... But no information on whether or not the center AF point functions with an f/8 lens like other 1-series bodies.
This excellent reference has detailed listings of which lens and f stop works with which of the 61 AF points and in what manner. There is zero mention of any lenses working with autofocus at f8. It appears that this feature which worked with all 1 series bodies in the past, has been removed with the 1D X.
The only good news is that I have learned that the slow clunky contrast-based autofocus available in LiveView works down to f11.3 on my T3i, so I assume this poor form of autofocus would work (probably better) on the 1D X.
Good pick up John, I don't want to start any rumors either, however, I would say that no AF @ f/8 on their flagship body would be a major oversight, to put it mildly.
That would mean that the 500mm f/4L or 600mm f/4L would not AF with a 2X extender, and this would also affect plenty of other combinations as well.
Lets hope that Canon did not screw this up.
Rich
A real "WTH" — it's one thing if the f/2.8 points work with an f/4 lens (combo), although I had even forgotten that from the 1D4 review. But that some or all f/2.8 points do NOT work with an f/2.8 lens? Won't that seriously affect AF precision? I remember that the f/2.8 points focus to within 1/3 of the f/2.8 DoF, the f/5.6 points only to within 1/2 of the f/5.6 DoF (what about the new f/4 points)?
You're welcome, I found the link in a thread about the 1DX in another forum, one where people still believe that the larger pixels were required for the higher ISO, a belief that this forum has mostly cured me of ;-)
From what I have read on 2 other forums, including an email from Chuck Westfall, as of right now the 1DX does not support AF at f/8 or higher!
I believe, however, that Canon has already heard from some of the premier nature photographers that it would be a good idea to remedy that if possible.
So maybe it will be available by the time the first production units are released in 6 months or so.
Colin, I don't want to re-hash the entire subject but... if ALL else is equal when you compare 2 sensors / camera systems .... the one with larger pixels should have better low light sensitivity simply because a larger pixel collects more photons and should have a higher signal to noise ratio.
The trouble is, it's almost impossible to eliminate ALL other variables in order to make the comparison valid between cameras
If you look at DxO Mark and their data analysis, even the medium format cameras don't beat the Nikon D3 sensor in terms of low light sensitivity so there is more to it than just sensor size and pixel number.
Joel, Thanks for the update. Lets hope so!
Rich
Colin, I don't want to re-hash the entire subject but... if ALL else is equal when you compare 2 sensors / camera systems .... the one with larger pixels should have better low light sensitivity simply because a larger pixel collects more photons and should have a higher signal to noise ratio.
But I believe to have understood that this is only true for the 100% view or per pixel, but if you compare two images with different resolutions such that each image "used the same total sensor area" then the quality will be the same...?
Posting Permissions
Reply With Quote